The preparation of valuable recombinant (genetically engineered) polypeptides, for example pharmaceutical proteins, relies frequently on techniques which involve the production of these polypeptides as fusion or hybrid proteins. These techniques are based upon the preparation of hybrid genes, i. e. genes comprising genetic material encoding the polypeptide of interest linked to genetic material additional to the gene of interest. Production of the fusion polypeptide involves the introduction of the hybrid gene into a biological host cell system, for example yeast cells, which permits the expression and accumulation of the fusion polypeptide. Recovery of the polypeptide of interest involves the performance of a cleavage reaction which results in the separation of the desired polypeptide from the “fusion partner”.
Despite the additional steps which are required to produce a protein of interest as a fusion protein, rather than directly in its active form, the production of hybrid proteins has been found to overcome a number of problems. Firstly, overproduced polypeptides can aggregate in the host cell in insoluble fractions known as inclusion bodies. Conversion of this insoluble material involves often slow and complex refolding methods, making protein purification difficult. Secondly, those proteins which are present in soluble form in the cytoplasm often are subject to degradation by host specific enzymes, thus reducing the amounts of active protein that can be recovered. Linking the polypeptide of interest to a fusion partner has been found to limit these problems. Fusion partners known to the prior art include maltose binding protein (Di Guan et al. (1988) Gene 67: 21-30), glutathione-S-transferase (Johnson (1989) Nature 338: 585-587), ubiquitin (Miller et al. (1989) Biotechnology 7: 698-704), β-galactosidase (Goeddel et al. (1979) Proc. Natl. Acad. Sci. (USA) 76: 106-110), and thioredoxin (LaVallie et al. (1993) Biotechnology 11:187-193).
It has also been proposed to employ fusion partners as affinity peptides. This methodology facilitates the isolation and recovery of the fusion peptide from the host cells by exploiting the physico-chemical properties of the fusion partner. (See, for example, WO 91/11454).
Finally, the use of a fusion partner may enable the production of a peptide which would otherwise be too small to accumulate and recover efficiently from a recombinant host cell system. This technology is described, for example, by Schultz et al., (1987, J. Bacteriol. 169: 5385-5392)
All of these procedures result in the production a hybrid protein in which the protein of interest is linked to an additional polypeptide. In order to recover the active polypeptide it is, in general, necessary to separate the fusion partner from the polypeptide of interest. Most commonly, a cleavage reaction, either by enzymatic or by chemical means, is performed. Such reactions employ agents that act by hydrolysis of peptide bonds and the specificity of the cleavage agent is determined by the identity of the amino acid residue at or near the peptide bond which is cleaved.
Enzymes known to the prior art as “proteolytic enzymes” have been found to be particularly well suited for the cleavage of fusion proteins. The cleavage reaction is performed by contacting the fusion protein with a proteolytic enzyme under appropriate conditions. An example of this methodology is described in U.S. Pat. No. 4,743,679 which discloses a process for the production of human epidermal growth factor comprising cleavage of a fusion protein by Staphylococcus aureus V8 protease.
By contrast, chemical cleavage involves the use of chemical agents which are known to permit hydrolysis of peptide bonds under specific conditions. Cyanogenbromide, for example, is known to cleave the polypeptide chain at a methionine residue. A hydrolysis reaction for the cyanogenbromide cleavage of the proteins urease and phosphorylase b based on this technique is described by Sekita et al. ((1975), Keio J. Med. 24: 203-210).
Both chemical and enzymatic cleavage reactions require the presence of a peptide bond which can be cleaved by the cleavage agent which is employed. For this reason it is often desirable to place an appropriate target sequence at the junction of the fusion partner and the target protein. Fusion peptides comprising “linker” sequences containing a target for a proteolytic enzyme may readily be constructed using conventional art-recognized genetic engineering techniques.
Despite their great utility, the prior art cleavage methods have been recognized to be either inefficient or lack cleavage specificity. Inefficient cleavage results in low protein purification efficiency, while the lack of cleavage specificity results in cleavage at several locations resulting in product loss and generation of contaminating fragments. This results frequently in the recovery of only a small fraction of the desired protein. In addition, the currently widely used proteolytic enzymes, such as blood clotting factor Xa and thrombin, are expensive, and contamination of final product with blood pathogens is a consideration.
In view of these shortcomings, the limitations of the cleavage methods known to the prior art are apparent.
Zymogens, such as pepsin and chymosin, are enzymes which are synthesized as inactive precursors in vivo. Under appropriate conditions, zymogens are activated to form the mature active protein in a process involving the cleavage of an amino-terminal peptide which can be referred to as the “pro-peptide”, “pro-region” or “pro-sequence”. Activation of zymogens may require the presence of an additional specific proteolytic enzyme, for example various hormones, such as insulin, are processed by a specific proteolytic enzyme. Alternatively, activation may occur without an additional enzymatic catalyst. These kinds of zymogens are frequently referred to as “autocatalytically maturing” zymogens. Examples of autocatalytically maturing zymogens include pepsin, pepsinogen and chymosin which are activated by an acidic environment, for example in the mammalian stomach.
The autocatalytic activation and processing of zymogens has been documented extensively (see for example, McCaman and Cummings, (1986), J. Biol. Chem. 261: 15345-15348; Koelsch et al. (1994). FEBS Letters 343: 6-10). It has also been documented that activation of the zymogen does not necessarily require a physical linkage of the pro-peptide to the mature protein (Silen et al. (1989), Nature, 341: 462-464).
There is a need for an improved process for recovering recombinantly produced polypeptides from their expression systems.